Following the publication this week of the research report in the Journal of Science on the success of the University of Alberta’s Deepstack poker bot last December (see previous report) the mainstream media have again given wide coverage to the 3,000-hand confrontation between a poker-playing computer program and 11 professional players.
InfoPowa readers may recall that the 11 players were recruited by the International Federation of Poker, who were asked to play Texas Hold’em against Deepstack over a period of four weeks…it didn’t end well for the humans.
The program, a product of collaboration between the University of Alberta, Charles University in Prague and Czech Technical University, was developed to use ‘intuition’ through deep learning – allowing it to reassess its strategy with each move and learn from the action.
Alberta U’s Professor Michael Bowling, reported:
“Poker has been a longstanding challenge problem in artificial intelligence. It is the quintessential game of imperfect information in the sense that the players don’t have the same information or share the same perspective while they’re playing.”
The professor claimed that DeepStack acts at a human speed – with an average of just three seconds of ‘thinking’ time – and runs on a simple gaming laptop with an Nvidia graphics processing unit.